I don't know, if I can ask this here, but I'm interested to know what kind of abstract data type (ADT) does Twitter use to model relations between profiles and why. I'm just starting to learn ADTs, so I would like to learn how they work in the real world as well as their applications.
Any platform, focused on social networking (not necessarily Twitter), at its core uses the most appropriate and natural abstract data type (ADT) for such domain - a graph data structure.
If you use Python, you can check nice NetworkX package, used for "the creation, manipulation, and study of the structure, dynamics, and functions of complex networks". Of course, there are many other software tools for various programming languages for building, using and analyzing network structures. You might also find useful the relevant book "Social Network Analysis for Startups: Finding connections on the social web", which provides a nice introduction into the social network analysis (SNA) and uses the above-mentioned NetworkX software for SNA examples. P.S. I have no affiliation whatsoever with NetworkX open source project or the book's authors.